Visual versus video methods for estimating reef fish biomass

S. K. Wilson, N. A.J. Graham, T. H. Holmes, M. A. MacNeil, N. M. Ryan

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Estimates of fish biomass collated at the community level are reliable indicators of fish and ecosystem health. Data to calculate fish biomass is routinely collected using either underwater visual census (UVC) or stereo diver operated video (DOV), although the compatibility of UVC and DOV based estimates are yet to be assessed. Accordingly, we calculated and compared community level measures of coral reef fish biomass at Ningaloo reef (Western Australia) using both UVC and DOV. The UVC based biomass estimates were 788 kg/Ha, which was ∼50% greater than those from DOV (500 kg/Ha). Differences between the methods were primarily due to DOV measuring the length of only ∼40% of fish detected by video, preventing fish specific weight calculations for all fish encountered. When the size of unmeasured fish was assumed to be the median value of fish measured by DOV, revised DOV+ estimates of community biomass (778 kg/Ha) were similar to those from UVC. However, even when unmeasured fish were included in DOV calculations, biomass of some families (serranids) were still higher when using UVC. Conversely, DOV adjusted estimates of pomacentrid biomass were higher than those from UVC, due to DOV measuring fewer small bodied fish (<3 cm), thus having a larger median size for the high number of unmeasured pomacentrids compared to UVC. Our results suggest that community measures of fish biomass from DOV and UVC are broadly comparable once weights of unmeasured fish are incorporated into DOV estimates. This may increase the spatial and temporal scales at which fish biomass can be monitored, although compatibility of data will depend on the composition and size distribution of the fish assemblages.

Original languageEnglish
Pages (from-to)146-152
Number of pages7
JournalEcological Indicators
Volume85
DOIs
Publication statusPublished - 1 Feb 2018

Fingerprint

reefs
reef
biomass
fish
census
methodology
method
video
Biomass
Fish
fish health
Census
ecosystem health
Western Australia
coral reefs
coral reef

Cite this

Wilson, S. K. ; Graham, N. A.J. ; Holmes, T. H. ; MacNeil, M. A. ; Ryan, N. M. / Visual versus video methods for estimating reef fish biomass. In: Ecological Indicators. 2018 ; Vol. 85. pp. 146-152.
@article{43f7422deb924fcaab5c28a1f5190a64,
title = "Visual versus video methods for estimating reef fish biomass",
abstract = "Estimates of fish biomass collated at the community level are reliable indicators of fish and ecosystem health. Data to calculate fish biomass is routinely collected using either underwater visual census (UVC) or stereo diver operated video (DOV), although the compatibility of UVC and DOV based estimates are yet to be assessed. Accordingly, we calculated and compared community level measures of coral reef fish biomass at Ningaloo reef (Western Australia) using both UVC and DOV. The UVC based biomass estimates were 788 kg/Ha, which was ∼50{\%} greater than those from DOV (500 kg/Ha). Differences between the methods were primarily due to DOV measuring the length of only ∼40{\%} of fish detected by video, preventing fish specific weight calculations for all fish encountered. When the size of unmeasured fish was assumed to be the median value of fish measured by DOV, revised DOV+ estimates of community biomass (778 kg/Ha) were similar to those from UVC. However, even when unmeasured fish were included in DOV calculations, biomass of some families (serranids) were still higher when using UVC. Conversely, DOV adjusted estimates of pomacentrid biomass were higher than those from UVC, due to DOV measuring fewer small bodied fish (<3 cm), thus having a larger median size for the high number of unmeasured pomacentrids compared to UVC. Our results suggest that community measures of fish biomass from DOV and UVC are broadly comparable once weights of unmeasured fish are incorporated into DOV estimates. This may increase the spatial and temporal scales at which fish biomass can be monitored, although compatibility of data will depend on the composition and size distribution of the fish assemblages.",
keywords = "Coral reef monitoring, Digital image techniques, Fisheries management, Methodological comparison, Size distribution",
author = "Wilson, {S. K.} and Graham, {N. A.J.} and Holmes, {T. H.} and MacNeil, {M. A.} and Ryan, {N. M.}",
year = "2018",
month = "2",
day = "1",
doi = "10.1016/j.ecolind.2017.10.038",
language = "English",
volume = "85",
pages = "146--152",
journal = "Ecological Indicators",
issn = "1470-160X",
publisher = "Pergamon",

}

Visual versus video methods for estimating reef fish biomass. / Wilson, S. K.; Graham, N. A.J.; Holmes, T. H.; MacNeil, M. A.; Ryan, N. M.

In: Ecological Indicators, Vol. 85, 01.02.2018, p. 146-152.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Visual versus video methods for estimating reef fish biomass

AU - Wilson, S. K.

AU - Graham, N. A.J.

AU - Holmes, T. H.

AU - MacNeil, M. A.

AU - Ryan, N. M.

PY - 2018/2/1

Y1 - 2018/2/1

N2 - Estimates of fish biomass collated at the community level are reliable indicators of fish and ecosystem health. Data to calculate fish biomass is routinely collected using either underwater visual census (UVC) or stereo diver operated video (DOV), although the compatibility of UVC and DOV based estimates are yet to be assessed. Accordingly, we calculated and compared community level measures of coral reef fish biomass at Ningaloo reef (Western Australia) using both UVC and DOV. The UVC based biomass estimates were 788 kg/Ha, which was ∼50% greater than those from DOV (500 kg/Ha). Differences between the methods were primarily due to DOV measuring the length of only ∼40% of fish detected by video, preventing fish specific weight calculations for all fish encountered. When the size of unmeasured fish was assumed to be the median value of fish measured by DOV, revised DOV+ estimates of community biomass (778 kg/Ha) were similar to those from UVC. However, even when unmeasured fish were included in DOV calculations, biomass of some families (serranids) were still higher when using UVC. Conversely, DOV adjusted estimates of pomacentrid biomass were higher than those from UVC, due to DOV measuring fewer small bodied fish (<3 cm), thus having a larger median size for the high number of unmeasured pomacentrids compared to UVC. Our results suggest that community measures of fish biomass from DOV and UVC are broadly comparable once weights of unmeasured fish are incorporated into DOV estimates. This may increase the spatial and temporal scales at which fish biomass can be monitored, although compatibility of data will depend on the composition and size distribution of the fish assemblages.

AB - Estimates of fish biomass collated at the community level are reliable indicators of fish and ecosystem health. Data to calculate fish biomass is routinely collected using either underwater visual census (UVC) or stereo diver operated video (DOV), although the compatibility of UVC and DOV based estimates are yet to be assessed. Accordingly, we calculated and compared community level measures of coral reef fish biomass at Ningaloo reef (Western Australia) using both UVC and DOV. The UVC based biomass estimates were 788 kg/Ha, which was ∼50% greater than those from DOV (500 kg/Ha). Differences between the methods were primarily due to DOV measuring the length of only ∼40% of fish detected by video, preventing fish specific weight calculations for all fish encountered. When the size of unmeasured fish was assumed to be the median value of fish measured by DOV, revised DOV+ estimates of community biomass (778 kg/Ha) were similar to those from UVC. However, even when unmeasured fish were included in DOV calculations, biomass of some families (serranids) were still higher when using UVC. Conversely, DOV adjusted estimates of pomacentrid biomass were higher than those from UVC, due to DOV measuring fewer small bodied fish (<3 cm), thus having a larger median size for the high number of unmeasured pomacentrids compared to UVC. Our results suggest that community measures of fish biomass from DOV and UVC are broadly comparable once weights of unmeasured fish are incorporated into DOV estimates. This may increase the spatial and temporal scales at which fish biomass can be monitored, although compatibility of data will depend on the composition and size distribution of the fish assemblages.

KW - Coral reef monitoring

KW - Digital image techniques

KW - Fisheries management

KW - Methodological comparison

KW - Size distribution

UR - http://www.scopus.com/inward/record.url?scp=85032002133&partnerID=8YFLogxK

U2 - 10.1016/j.ecolind.2017.10.038

DO - 10.1016/j.ecolind.2017.10.038

M3 - Article

VL - 85

SP - 146

EP - 152

JO - Ecological Indicators

JF - Ecological Indicators

SN - 1470-160X

ER -